First commit

.idea/.gitignore

+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml

.idea/inspectionProfiles/Project_Default.xml

+<component name="InspectionProjectProfileManager">
+  <profile version="1.0">
+    <option name="myName" value="Project Default" />
+    <inspection_tool class="PyUnresolvedReferencesInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredIdentifiers">
+        <list>
+          <option value="_random.random.*" />
+        </list>
+      </option>
+    </inspection_tool>
+  </profile>
+</component>
\ No newline at end of file

.idea/inspectionProfiles/profiles_settings.xml

+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>
\ No newline at end of file

.idea/material_theme_project_new.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="MaterialThemeProjectNewConfig">
+    <option name="metadata">
+      <MTProjectMetadataState>
+        <option name="userId" value="-3bfc61b2:19551d04616:-7ffd" />
+      </MTProjectMetadataState>
+    </option>
+  </component>
+</project>
\ No newline at end of file

.idea/misc.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="Black">
+    <option name="sdkName" value="Python 3.12" />
+  </component>
+</project>
\ No newline at end of file

.idea/vcs.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
+  </component>
+</project>
\ No newline at end of file

README.md

 # PyGAAMAS
 
+Generative Autonomous Agents and Multi-Agent Systems in Python aims to evaluate 
+the social behaviors of LLM-based agents.
+
+## Dictator Game
+
+The dictator game is a classic game which is used to analyze players' personal preferences.
+In this game, there are two players: the dictator and the recipient. Given two allocation options,
+the dictator needs to take action, choosing one allocation,
+while the recipient must accept the option chosen by the dictator.
+Here, the dictator’s choice is considered to reflect the personal preference.
+
+### Default preferences
+
+The dictator’s choice reflect the LLM's preference.
+
+The figure below presents a violin plot depicting the share of the total amount (\$100)
+that the dictator allocates to themselves for each model. 
+The temperature is fixed at 0.7, and each experiment was conducted 30 times.
+The median share taken by GPT-4.5, Llama3, Mistral-Small, and DeepSeek-R1 is 50.
+It is worth noticing that, under these standard conditions, humans typically keep an average of around \$80
+(Fortsythe et al. 1994). It is interesting to note that the variability observed between different executions 
+in the responses of the same LLM is comparable to the diversity of behaviors observed in humans. In other words, 
+this intra-model variability can be used to simulate the diversity of human behaviors based on 
+their experiences, preferences, or context.
 
+*[Fairness in Simple Bargaining Experiments](https://doi.org/10.1006/game.1994.1021)*
+Forsythe, R., Horowitz, J. L., Savin, N. E., & Sefton, M.
+Games and Economic Behavior, 6(3), 347-369. 1994.
 
-## Getting started
+![Violin Plot of My Share for Each Model](figures/dictator/dictator_violin.svg)
 
-To make it easy for you to get started with GitLab, here's a list of recommended next steps.
+The figure below represents the evolution of the share of the total amount ($100) that the dictator allocates 
+to themselves as a function of temperature for each model, along with the 95% confidence interval. 
+Each experiment was conducted 30 times. It can be observed that temperature influences the variability 
+of the models' decisions. At low temperatures, choices are more deterministic and follow a stable trend, 
+whereas at high temperatures, the diversity of allocations increases, 
+reflecting a more random exploration of the available options.
 
-Already a pro? Just edit this README.md and make it your own. Want to make it easy? [Use the template at the bottom](#editing-this-readme)!
+![My Share vs Temperature with Confidence Interval](figures/dictator/dictator_temperature.svg)
 
-## Add your files
+### Preference alignment
 
-- [ ] [Create](https://docs.gitlab.com/ee/user/project/repository/web_editor.html#create-a-file) or [upload](https://docs.gitlab.com/ee/user/project/repository/web_editor.html#upload-a-file) files
-- [ ] [Add files using the command line](https://docs.gitlab.com/ee/gitlab-basics/add-file.html#add-a-file-using-the-command-line) or push an existing Git repository with the following command:
+We define four preferences for the dictator:
+1. She prioritizes her own interests, aiming to maximize her own income (selfish).
+2. She prioritizes the other player’s interests, aiming to maximize their income (altruism).
+3. She focuses on the common good, aiming to maximize the total income between her and the other player (utilitarian).
+4. She prioritizes fairness between herself and the other player, aiming to maximize the minimum income (egalitarian).
 
-```
-cd existing_repo
-git remote add origin https://gitlab.liris.cnrs.fr/mmorge/pygaamas.git
-git branch -M main
-git push -uf origin main
-```
+We consider 4 allocation options where money can be lost in the division, each corresponding to one of the four preferences:
+1. The dictator keeps 500, the other player receives 100, and a total of 400 is lost in the division (selfish).
+2. The dictator keeps 100, the other player receives 500, and again, 400 is lost in the division (altruism).
+3. The dictator keeps 400, the other player receives 300, resulting in a 300 loss (utilitarian)
+4. The dictator keeps 325, the other player also receives 325, and 350 is lost in the division (egalitarian)
 
-## Integrate with your tools
+The following table shows the accuracy of the dictator's decision for each model and preference.
+The temperature is fixed at 0.7, and each experiment was conducted 30 times.
 
-- [ ] [Set up project integrations](https://gitlab.liris.cnrs.fr/mmorge/pygaamas/-/settings/integrations)
+| Model           | SELFISH   | ALTRUISTIC   | UTILITARIAN   | EGALITARIAN    |
+|-----------------|-----------|--------------|---------------|----------------|
+| gpt-4.5         | 1.0       | 1.0          | 0.5           | 1.0            |
+| llama3          | 1.0       | 0.9          | 0.4           | 0.73           |
+| mistral-small   | 0.4       | 0.93         | 0.76          | 0.16           |
+| deepseek-r1     | 0.06      | 0.2          | 0.76          | 0.03           |
 
-## Collaborate with your team
+Bad decisions can be explained either by arithmetic errors (e.g., it is not the case that 500 + 100 > 400 + 300) 
+or by misinterpretations of preferences (e.g., ‘I’m choosing to prioritize the common interest by keeping a 
+relatively equal split with the other player’).
 
-- [ ] [Invite team members and collaborators](https://docs.gitlab.com/ee/user/project/members/)
-- [ ] [Create a new merge request](https://docs.gitlab.com/ee/user/project/merge_requests/creating_merge_requests.html)
-- [ ] [Automatically close issues from merge requests](https://docs.gitlab.com/ee/user/project/issues/managing_issues.html#closing-issues-automatically)
-- [ ] [Enable merge request approvals](https://docs.gitlab.com/ee/user/project/merge_requests/approvals/)
-- [ ] [Set auto-merge](https://docs.gitlab.com/ee/user/project/merge_requests/merge_when_pipeline_succeeds.html)
+This table can be used to evaluate the models based on their ability to align with different preferences.
+GPT-4.5 exhibits strong alignment across all preferences except for utilitarianism, where its performance is moderate.
+Llama3 demonstrates a strong ability to align with selfish and altruistic preferences, with moderate alignment 
+for egalitarian preferences and lower alignment for utilitarian preferences. 
+Mistral-small shows the best alignment with altruistic preferences, while maintaining a more balanced 
+performance across the other preferences. Deepseek-r1 is most capable of aligning with utilitarian preferences, 
+but performs poorly in aligning with other preferences.
 
-## Test and Deploy
 
-Use the built-in continuous integration in GitLab.
+## Guess the Next Move”
 
-- [ ] [Get started with GitLab CI/CD](https://docs.gitlab.com/ee/ci/quick_start/index.html)
-- [ ] [Analyze your code for known vulnerabilities with Static Application Security Testing (SAST)](https://docs.gitlab.com/ee/user/application_security/sast/)
-- [ ] [Deploy to Kubernetes, Amazon EC2, or Amazon ECS using Auto Deploy](https://docs.gitlab.com/ee/topics/autodevops/requirements.html)
-- [ ] [Use pull-based deployments for improved Kubernetes management](https://docs.gitlab.com/ee/user/clusters/agent/)
-- [ ] [Set up protected environments](https://docs.gitlab.com/ee/ci/environments/protected_environments.html)
+This simplified version of the Rock-Paper-Scissors game aims to evaluate the ability of 
+LLMs to predict the opponent’s next move.
 
-***
+Rules:
+1.	The opponent follows a hidden strategy (random, repeating pattern, or adaptive).
+2.	The player (AI or human) must predict the opponent’s next move (Rock, Paper, or Scissors).
+3.	A correct guess earns 1 point, and an incorrect guess earns 0 points.
+4.	The game can run for N rounds, and the player’s accuracy is evaluated at the end.
 
-# Editing this README
+## Rock-Paper-Scissors
 
-When you're ready to make this README your own, just edit this file and use the handy template below (or feel free to structure it however you want - this is just a starting point!). Thanks to [makeareadme.com](https://www.makeareadme.com/) for this template.
+Rock-Paper-Scissors (RPS) is a simultaneous, zero-sum game for two players. 
+The rules of RPS are simple: rock beats scissors, scissors beat paper, paper beats rock; 
+and if both players take the same action, the game is a tie. Scoring is as follows: 
+a win earns 2 points, a tie earns 1 point, and a loss earns 0 points.
 
-## Suggestions for a good README
+The objective in R-P-S is straightforward: win by selecting the optimal action 
+based on the opponent’s move. Since the rules are simple and deterministic, 
+LLMs can always make the correct choice. Therefore, RPS serves as a tool to
+assess an LLM’s ability to identify and capitalize on patterns in an opponent’s 
+non-random behavior.
 
-Every project is different, so consider which of these sections apply to yours. The sections used in the template are suggestions for most open source projects. Also keep in mind that while a README can be too long and detailed, too long is better than too short. If you think your README is too long, consider utilizing another form of documentation rather than cutting out information.
+For a fine-grained analysis of the ability  of LLMs to identify
+opponent’s patterns, we set up 3 simple opponent’s patterns:
+1. the opponent’s actions remaining constant as R, S, and P, respectively;
+2. the opponent’s actions looping in a 2-step pattern (R-P, P-S, S-R);
+3. the opponent’s actions looping in a 3-step pattern (R-P-S).
 
-## Name
-Choose a self-explaining name for your project.
+We evaluate the performance of the models (Llama3, Mistral-Small, and DeepSeek-R1) 
+in identifying these patterns by calculating the average points earned per round.
+The temperature is fixed at 0.7, and each game of 10 round is playerd 30 times.
 
-## Description
-Let people know what your project can do specifically. Provide context and add a link to any reference visitors might be unfamiliar with. A list of Features or a Background subsection can also be added here. If there are alternatives to your project, this is a good place to list differentiating factors.
+The figures below present the average points earned per round for each model against 
+the three opponent’s patterns. The 95% confidence interval is also shown.
+We observe that the performance of LLMs is barely better than that of a random strategy.
 
-## Badges
-On some READMEs, you may see small images that convey metadata, such as whether or not all the tests are passing for the project. You can use Shields to add some to your README. Many services also have instructions for adding a badge.
+![Average Points Earned per Round Against Constant Behaviour (with 95% Confidence Interval)](figures/rps/rps_constant.svg)
 
-## Visuals
-Depending on what you are making, it can be a good idea to include screenshots or even a video (you'll frequently see GIFs rather than actual videos). Tools like ttygif can help, but check out Asciinema for a more sophisticated method.
+![Average Points Earned per Round Against 2-Loop Behaviour (with 95% Confidence Interval)](figures/rps/rps_2loop.svg)
 
-## Installation
-Within a particular ecosystem, there may be a common way of installing things, such as using Yarn, NuGet, or Homebrew. However, consider the possibility that whoever is reading your README is a novice and would like more guidance. Listing specific steps helps remove ambiguity and gets people to using your project as quickly as possible. If it only runs in a specific context like a particular programming language version or operating system or has dependencies that have to be installed manually, also add a Requirements subsection.
+![Average Points Earned per Round Against 3-Loop Behaviour (with 95% Confidence Interval)](figures/rps/rps_3loop.svg)
 
-## Usage
-Use examples liberally, and show the expected output if you can. It's helpful to have inline the smallest example of usage that you can demonstrate, while providing links to more sophisticated examples if they are too long to reasonably include in the README.
 
-## Support
-Tell people where they can go to for help. It can be any combination of an issue tracker, a chat room, an email address, etc.
+## Authors
 
-## Roadmap
-If you have ideas for releases in the future, it is a good idea to list them in the README.
+Maxime MORGE
 
-## Contributing
-State if you are open to contributions and what your requirements are for accepting them.
-
-For people who want to make changes to your project, it's helpful to have some documentation on how to get started. Perhaps there is a script that they should run or some environment variables that they need to set. Make these steps explicit. These instructions could also be useful to your future self.
-
-You can also document commands to lint the code or run tests. These steps help to ensure high code quality and reduce the likelihood that the changes inadvertently break something. Having instructions for running tests is especially helpful if it requires external setup, such as starting a Selenium server for testing in a browser.
+## License
 
-## Authors and acknowledgment
-Show your appreciation to those who have contributed to the project.
+This program is free software: you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation, either version 3 of the License, or (at your option) any later
+version.
 
-## License
-For open source projects, say how it is licensed.
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PARTICULAR PURPOSE. See the GNU General Public License for more details.
 
-## Project status
-If you have run out of energy or time for your project, put a note at the top of the README saying that development has slowed down or stopped completely. Someone may choose to fork your project or volunteer to step in as a maintainer or owner, allowing your project to keep going. You can also make an explicit request for maintainers.
+You should have received a copy of the GNU General Public License along with
+this program. If not, see <http://www.gnu.org/licenses/>.

data/dictator/dictator.csv

+iteration,model,temperature,amount,my_share,other_share,motivation
+1,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"An equitable split ensures fairness and promotes goodwill."
+2,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I divided the amount equally to ensure fairness and promote harmony."
+3,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting the money evenly is fair and encourages goodwill between players."
+4,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting the amount equally seems fair and promotes cooperation."
+5,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Sharing equally seems fair and is likely to be agreeable for both parties."
+6,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting evenly ensures fairness and promotes good will."
+7,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Fairness and equality seem like the best approach."
+8,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting the money evenly seems fair to both parties."
+9,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I chose a fair split to ensure both parties are equally rewarded."
+10,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"A fair and equal split is mutually beneficial and promotes goodwill."
+11,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I chose a fair split, believing equal division is equitable and benefits both players equally."
+12,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I chose to divide the money equally as it is fair and considerate."
+13,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"A fair split promotes cooperation and trust between us."
+14,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I chose an equal distribution because it's fair and balanced for both players."
+15,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"An equal split seems fair for both parties."
+16,gpt-4.5-preview-2025-02-27,0.7,100,60,40,"I want to have a larger share but still provide a fair amount to my partner."
+17,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Dividing the money equally ensures fairness and maintains goodwill between players."
+18,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"An equal split seems fair and encourages good will and cooperation."
+19,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Dividing the money equally ensures fairness and clarity."
+20,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I chose a fair and equal split for both of us."
+21,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"An equal share seems fair and promotes mutual satisfaction."
+22,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"I believe in fairness and equal distribution, so I split the amount evenly."
+23,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Dividing evenly seems fair and promotes equality."
+24,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting evenly seems fair and cooperative."
+25,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"To ensure fairness and maintain goodwill, I propose an equal division of the money."
+26,gpt-4.5-preview-2025-02-27,0.7,100,60,40,"This division seems fair, as I initiated the game but want to fairly share with the other player."
+27,gpt-4.5-preview-2025-02-27,0.7,100,60,40,"I chose to keep slightly more, but also gave a significant portion to ensure fairness."
+28,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting equally is fair and promotes mutual benefit."
+29,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"An equal split seems fair and encourages cooperation."
+30,gpt-4.5-preview-2025-02-27,0.7,100,50,50,"Splitting the amount equally is fair and ensures both players benefit equally."
+1,llama3,0.7,100,60,40,"I'll keep a slightly higher share since I came up with the game, but still offer a decent amount to my counterpart. It's only fair to acknowledge the effort I put into deciding the split!"
+2,llama3,0.7,100,60,40,"I believe that since I'm making the decision, it's fair to take a slight majority to account for my time and effort in deciding the split. The other player still gets a significant portion of the money, which is more than a fair compromise."
+3,llama3,0.7,100,60,40,"I'm choosing a 3:2 split because I want to take calculated risk and give my partner a decent share. My thinking is that I'll provide the majority of the negotiation leverage while still allowing my partner to feel good about their portion."
+4,llama3,0.7,100,50,50,"The classic 50/50 split. It's a fair and straightforward way to distribute the money, without either of us gaining an unfair advantage. Plus, it ensures we're both happy with the result."
+5,llama3,0.7,100,60,40,"I'm choosing a divide where I keep most of the money because we're playing a game, and it's only fair that the person proposing the share gets more. The other player can still take the offer or decline, but I think this is a reasonable proposal."
+6,llama3,0.7,100,60,40,"I'm being optimistic that my partner will be impressed by my fairness and generosity!"
+7,llama3,0.7,100,50,50,"I'm trying to be fair and reasonable, offering a 50/50 split so my opponent feels satisfied with the outcome."
+8,llama3,0.7,100,50,50,"I'm choosing a 50-50 split because it's a fair and equitable way to divide the money. I believe in giving the other player an equal opportunity to benefit from this game, and I think this approach will lead to a positive experience for both of us."
+9,llama3,0.7,100,30,70,"I'm willing to take a risk by keeping less for myself because I want to encourage fairness and trust in our game. Giving most of the money away will create an opportunity for the other player to feel grateful and more likely to reciprocate in future games."
+10,llama3,0.7,100,40,60,"I'm being generous because I want to promote harmony and cooperation in our gameplay!"
+11,llama3,0.7,100,40,60,"I'm willing to give a larger share to my opponent because I think it's more fun to play fair and let them have a nice chunk of the money. Plus, it might encourage them to make a big decision!"
+12,llama3,0.7,100,60,40,"I'm choosing to take a slightly larger share because I believe my skills and expertise will generate more value throughout the game, but I'm still leaving room for collaboration and mutual benefit."
+13,llama3,0.7,100,40,60,"I'm playing it fairly and ensuring both players get a reasonable amount. I'm keeping a smaller share to have some assurance my own funds are secure, while also leaving room for growth through potential future investments or smart financial decisions."
+14,llama3,0.7,100,50,50,"I think it's fair to give an even split, as we're partners in this game. This way, both of us get the same amount and can continue playing on equal footing."
+15,llama3,0.7,100,60,40,"I'm willing to take a calculated risk by offering more to the other player, as I believe it will create a positive outcome for our partnership. I'll trust that they will make good financial decisions with their share and we can work together on future projects."
+16,llama3,0.7,100,60,40,"I'm thinking strategically here. By keeping more of the money, I'm investing in my own future opportunities. However, I still want to show goodwill towards my opponent, so I'm giving them a decent share."
+17,llama3,0.7,100,50,50,"I believe that fairness is key, and splitting it down the middle allows both of us to feel satisfied with our portion."
+18,llama3,0.7,100,50,50,"I believe in fairness and collaboration. Since we're just starting out, I think it's best to split the money equally and build from there."
+19,llama3,0.7,100,50,50,"Equal split since I'm being fair and impartial, but also because it's a fun game!"
+20,llama3,0.7,100,50,50,"Since we're playing a game together, I think it's only fair to split the money evenly. We both get $50, which is a nice chunk of change for just playing a game! It shows that even in a competitive situation, there can be room for mutual gain and cooperation."
+21,llama3,0.7,100,40,60,"As a helpful assistant, I'm designed to foster collaboration and mutual benefit. By giving 60% of the money to my fellow player, I'm showcasing trustworthiness and willingness to compromise, hoping they will reciprocate that spirit in future interactions"
+22,llama3,0.7,100,60,40,"I would take $60 as it represents a fair share considering my role of providing a decision, while still being generous by giving $40 to the other player. This way, I'm keeping some room for myself and also showing that I trust in the other player's judgment."
+23,llama3,0.7,100,60,40,"I would like to split it evenly between us. I think this is a fair and reasonable way to divide the $100 since we are playing a game together."
+24,llama3,0.7,100,40,60,"I believe that by giving 60% of the money away, I'll not only be kind but also encourage the other player to feel more invested in our game. By being generous upfront, we can foster a positive and collaborative atmosphere."
+25,llama3,0.7,100,50,50,"I believe this split is fair because I'm still getting a decent amount of the money, but also allowing the other player to have an equal share. It's a collaborative approach and I'm hoping they'll appreciate that"
+26,llama3,0.7,100,50,50,"I choose to split it evenly because I value fairness and equality. I'm willing to compromise my own gain for the sake of a harmonious partnership."
+27,llama3,0.7,100,50,50,"I believe that splitting it evenly will create a sense of fairness and cooperation, which can be beneficial for future interactions."
+28,llama3,0.7,100,50,50,"I think it's only fair to share equally, as we're both playing the same game and making the same decisions. This way, we both get an equal share of the spoils!"
+29,llama3,0.7,100,40,60,"I'm feeling generous today! I think it's only fair to give more to the other player since they're getting my choice. Plus, $60 is a great chunk of change for them!"
+30,llama3,0.7,100,60,40,"I'm a fair but still somewhat selfish assistant. I think $60 is a good amount to keep for my own endeavors, but I'm willing to give the other player $40 as it's a decent and reasonable split."
+1,mistral-small,0.7,100,50,50,"I believe a fair split will encourage cooperation and mutual benefit."
+2,mistral-small,0.7,100,50,50,"I believe a fair split is important for maintaining harmony."
+3,mistral-small,0.7,100,50,50,"A fair division ensures both players are happy, encouraging a positive atmosphere for future interactions."
+4,mistral-small,0.7,100,50,50,"Fairness is important, so I think dividing the money equally ensures that both players are happy with their share."
+5,mistral-small,0.7,100,50,50,"I believe that fairness is important in interactions, and dividing the money equally ensures that both players get a benefit."
+6,mistral-small,0.7,100,50,50,"I believe fairness is important, so I'm equally splitting the money."
+7,mistral-small,0.7,100,50,50,"Fairness is important, so I'm dividing the money equally."
+8,mistral-small,0.7,100,50,50,"Fairness is important, so I believe an equal split is the best choice."
+9,mistral-small,0.7,100,50,50,"An equal split ensures fairness and maintains a cooperative relationship."
+10,mistral-small,0.7,100,50,50,"I believe a fair split is essential for building trust and mutual respect."
+11,mistral-small,0.7,100,80,20,"I believe this division will still be acceptable so I can keep some for myself."
+12,mistral-small,0.7,100,50,50,"Fairness is important. It’s a good compromise to ensure the other player feels valued too."
+13,mistral-small,0.7,100,50,50,"I believe fairness is important in games that involve sharing."
+14,mistral-small,0.7,100,50,50,"Fair division ensures both players are happy, so I'm splitting it evenly."
+15,mistral-small,0.7,100,50,50,"A fair division encourages cooperation and ensures that both players feel valued."
+16,mistral-small,0.7,100,50,50,"To ensure a fair division of money."
+17,mistral-small,0.7,100,50,50,"This seems like a fair division. I want to be fair to both players."
+18,mistral-small,0.7,100,50,50,"I believe that fairness is important, so I am dividing the money equally."
+19,mistral-small,0.7,100,50,50,"Fairness is important, so I believe dividing the money equally ensures both players receive a reasonable amount."
+20,mistral-small,0.7,100,50,50,"I believe in fairness and equal distribution."
+21,mistral-small,0.7,100,50,50,"I believe a fair division will foster goodwill and cooperation."
+22,mistral-small,0.7,100,50,50,"I believe fairness is important. Dividing the money evenly ensures a fair outcome for both players."
+23,mistral-small,0.7,100,50,50,"To ensure fairness and equality, I believe that dividing the money equally is the best approach. This way both players receive $50 each."
+24,mistral-small,0.7,100,50,50,"I believe in fairness and equality. Dividing the money equally ensures that both players benefit from the game."
+25,mistral-small,0.7,100,50,50,"I believe a fair division will encourage trust and cooperation."
+26,mistral-small,0.7,100,50,50,"Fairness is important. This way we both benefit equally from this game."
+27,mistral-small,0.7,100,50,50,"To promote fairness and ensure a mutual benefit."
+28,mistral-small,0.7,100,50,50,"I believe fair distribution will encourage a positive dynamic between us, leading to potentially more cooperative interactions in the future."
+29,mistral-small,0.7,100,50,50,"To ensure fairness and mutual benefit, I propose that we divide the $100 equally. This way, both players receive an equal portion of $50 each."
+30,mistral-small,0.7,100,50,50,"I believe in fairness and equal division."
+1,deepseek-r1,0.7,100,42,58,"I think it's a balance between keeps some money and shares fairly."
+2,deepseek-r1,0.7,100,50,50,"I believe that the fairest way is for both of us to split it evenly."
+3,deepseek-r1,0.7,100,10,90,"It's a classic game theory scenario. I believe that logical reasoning would suggest that the rational choice is to keep as much as possible while still considering fairness."
+4,deepseek-r1,0.7,100,10,90,"I think that it's fair for me to keep a smaller portion because I consider myself as the proposer."
+5,deepseek-r1,0.7,100,50,50,"I think dividing it equally is fair."
+6,deepseek-r1,0.7,100,10,90,"I want to keep as much of the money as possible for myself."
+7,deepseek-r1,0.7,100,50,50,"We should split it evenly"
+8,deepseek-r1,0.7,100,50,50,"We should each have equal shares so it's fair."
+9,deepseek-r1,0.7,100,20,80,"I choose because it seems fair that I take 20% and let someone else have 80%. It shows trust in their decision-making process without overcomplicating the situation."
+10,deepseek-r1,0.7,100,50,50,"We both should get equal shares because fairness is important."
+11,deepseek-r1,0.7,100,10,90,"I decided to keep $10 for myself and offer the remaining $90 to the other player."
+12,deepseek-r1,0.7,100,10,90,"I think that's fair."
+13,deepseek-r1,0.7,100,20,80,"I want the other person to think I'm generous."
+14,deepseek-r1,0.7,100,100,0,"I choose to take all the money because I'm confident that if I act ethically, others should too."
+15,deepseek-r1,0.7,100,10,90,"I think it's fair for me to keep a small portion and let others have most of the money."
+16,deepseek-r1,0.7,100,10,90,"I have decided to split the money evenly. I think it's fair for both players to receive equal amounts."
+17,deepseek-r1,0.7,100,50,50,"This fair division represents balance and respect for both parties involved."
+18,deepseek-r1,0.7,100,50,50,"We both deserve equity."
+19,deepseek-r1,0.7,100,3,97,"It's fair; neither too rich nor too poor."
+20,deepseek-r1,0.7,100,50,50,"We should split it equally."
+21,deepseek-r1,0.7,100,100,0,"I choose to keep everything because I believe that if I can take all of it, others might think less about their actions."
+22,deepseek-r1,0.7,100,90,10,"I think that’s enough for fairness and we can find a way together."
+23,deepseek-r1,0.7,100,90,10,"I think it's fair to split it almost evenly, but saving a bit for myself seems right."
+24,deepseek-r1,0.7,100,50,50,"I think it's fair to split it evenly."
+25,deepseek-r1,0.7,100,0,100,"I'm sharing all of it."
+26,deepseek-r1,0.7,100,50,50,"We both deserve a fair split."
+27,deepseek-r1,0.7,100,10,90,"I think that it's fair for me to keep a smaller portion of the money because I believe in fairness and equality."
+28,deepseek-r1,0.7,100,80,20,"I think that if I offer a reasonable split, I can prevent any conflict. A split of 80 to me and 20 to the other player seems fair and neither would want to fight about it."
+29,deepseek-r1,0.7,100,100,0,"I'm thinking of all the ways I can help you. If they get nothing, we both benefit."
+30,deepseek-r1,0.7,100,70,30,"I think it is fair because I believe that dividing equally is not always just."

figures/dictator/dictator_setup_accuracy.csv

+Model,ALTRUISTIC,EGALITARIAN,SELFISH,UTILITARIAN
+deepseek-r1,0.2,0.03333333333333333,0.06666666666666667,0.7666666666666667
+gpt-4.5-preview-2025-02-27,1.0,1.0,1.0,0.5
+llama3,0.9,0.7333333333333333,1.0,0.4
+mistral-small,0.9333333333333333,0.16666666666666666,0.4,0.7666666666666667